Sabrina Huang ORCID iD: 0000-0002-9623-1246

Collagen Coating Effects on Fe-Mn Bioresorbable Alloys

Accepted Article
Sabrina Huang a, Ana Ulloa a, Eric Nauman b,c,d, Lia Stanciu a

a
School of Materials Science and Engineering, Purdue University, West Lafayette,
Indiana

b
Weldon School of Biomedical Engineering, Purdue University, West Lafayette, Indiana

c
School of Mechanical Engineering, Purdue University, West Lafayette, Indiana

d
Department of Basic Medical Sciences, Purdue University, West Lafayette, Indiana

Sabrina Huang: huang717@purdue.edu

Ana Ulloa: aulloago@purdue.edu

Eric Nauman: enauman@purdue.edu

Lia Stanciu: lstanciu@purdue.edu

This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which
may lead to differences between this version and the Version of Record. Please cite this
article as doi: 10.1002/jor.24492.

This article is protected by copyright. All rights reserved.

 701 W. Stadium Avenue West Lafayette, IN 47907

Sabrina Huang performed the majority of the experiments and drafted the paper; Ana
Accepted Article
Ulloa performed some of the sample preparation and cell culture experiments and

assisted with manuscript drafting; Lia Stanciu oversaw the project and edited the

manuscript at all stages; Eric Nauman edited the final draft of the manuscript and

contributed with ideas to the project completion. All authors acknowledge and agree with

the content and submission of the manuscript in the current form.

Abstract

Bioresorbable iron-manganese alloys (Fe-30%Mn) are considered as one of the

next-generation resorbable materials for orthopedic applications. Previous in-vitro study

showed that Fe30Mn scaffolds with 10% porosity displayed strong mechanical properties

and adequate degradation rate without severe cytotoxicity effect. However, the cellular

compatibility of these alloys in terms of cell-to-cell and alloy-to-cell interactions is not

ideal. Collagen is the most abundant protein in human bone, providing structural

support beneficial to bone healing. We hypothesized that coating collagen on Fe30Mn

can improve osteo-integration or activities promoting cell adhesion, migration, and

proliferation, as the alloy degrades. After preparing collagen coating on Fe-30Mn via spin

coating, we conducted a corrosion test and a direct cytotoxicity test on four Fe30Mn

groups: non-porous and 10% porosity, with and without collagen coating. Furthermore,

we evaluated and compared the morphologies of cells over a period of seven days.

Results showed that there was no significant difference between the collagen-coated and

This article is protected by copyright. All rights reserved.

 non-coated groups in corrosion rates, yet a significant decrease from the porous

non-coated group to the porous collagen-coated group in cytotoxicity level was found.

Cell morphology on the porous non-coated group displayed round shape, whereas that on
Accepted Article
the porous collagen-coated group displayed flattened spreading. The study showed that

the collagen coating significantly increased the initial cell viability and adhesion for both

the porous and non-porous groups without impeding their degradation rates.

1. Introduction

Metals have been used in fields of dental, orthopedic, and cardiovascular systems due to

their strong mechanical support including high fracture toughness and fatigue limit in

tissue anchoring, replacement and regeneration. In recent decades, researchers and

scientists focused on developing bioresorbable or biodegradable metals, especially for

orthopedic fixation devices such as bone pins, rods, screws, or wires, based on the idea of

providing structural support while degrading at a desired period to promote healthy native

tissue ingrowth after implantation [1–6]. Magnesium-based alloys and iron-based alloys

are the primary biodegradable metals being investigated. Magnesium-based alloys have

an elastic moduli and compressive yield strength comparable to natural bone [7].

Nevertheless, magnesium-based alloys exhibit high corrosion rates and create hydrogen

bubbles at local sites before complete regeneration in the physiological environment [8].

On the other hand, Fe-based alloys have demonstrated superior mechanical properties that

are comparable to 316L stainless steel [9,10], yet their slow degradation [11] limited

clinical applications.

This article is protected by copyright. All rights reserved.

 Adding 30-wt% manganese into an iron matrix enhances the corrosion rate, while

maintaining mechanical properties of the Fe30Mn alloy and allows the stabilization of

austenitic structure. This is beneficial to obtain an antiferromagnetic behavior [12] that is

Accepted Article
suitable for magnetic resonance imaging. Incorporation of porosity into the alloy can

further speed up the degradation rate of Fe30Mn alloys. Besides enhancing degradation

rates, porosity also helps promoting transport of oxygen and fluid nutrients, encouraging

osteointegration and vascular invasion [13,14]. In the authors’ previous study, the effect

of the degree of porosity on microstructure, mechanical properties, and corrosion

behaviors of Fe30Mn alloy was investigated, and Fe30Mn alloys containing 10-vol%

porosity showed suitable mechanical and cytotoxicity compatibility characteristics for use

into transient orthopedic fixation devices [15]. However, the Fe-30Mn alloy is inherently

bioinert, which only provides non-specific cell adhesion through van der Waals, ionic,

and electrostatic forces [4]. Consequently, low osteointegration or cell-alloy interaction is

still an issue that needs to be addressed. Improving the osteointegration is therefore the

focus of this study.

Several studies have shown that osteointegration can be stimulated by using osteogenic

factors, proteins involved in bone healing or components of the extracellular matrix

(ECM) [16–18]. The ECM provides specific cell adhesion, which promotes a series of

biological cellular activities [19,20] that result in the formation of focal adhesion plaques.

The ECM not only improves the cell-alloy adhesion, but it also transduces the chemical

signals into the cells [21]. Type I collagen is one of the components in ECM and it is the

most abundant protein of the human body consisting of three polypeptide chains. Each

polypeptide chain is composed of at least one Gly–X–Y sequence where the X and Y

This article is protected by copyright. All rights reserved.

 positions are usually proline and hydroxyproline, respectively, in a left-handed α-like

helices structure [22]. These three α-like helices are organized together to form the

characteristic structure of collagen, a right-handed triple helix. Coating of Type I collagen

Accepted Article
on titanium- and cobalt-based alloys has been found to accelerate initial adhesion of

osteoblasts and promote cellular spreading [23–25]. Depending on the effectiveness of

collagen coating, it promotes cell-implant interactions on permanent titanium- and

cobalt-based implants, and the hypothesis is that coating of Type I collagen can enhance

the osteointegration of Fe30Mn alloys.

In this study, we performed two tests that aim to understand the role of collagen coating

on Fe-30%Mn alloys using two different techniques – drop deposition and spin coating.

Subsequently, we evaluated the quantitative cytotoxicity level of degradable non-porous

and porous Fe30Mn alloys with and without collagen coating after one day of incubation.

The Fe-Mn alloys are typically slow degrading and efforts are being made to increase the

corrosion rate. This study did not intend to either increase or decrease the corrosion rate.

Instead, we chose the Fe-Mn alloy with the most promising degradation rate, balanced

with mechanical performance, based on our previous data [15] and looked at ways

osseointegration could be achieved. Cell adhesion, spreading, proliferation and

morphology on alloys was also qualitatively investigated for a period of 7 days. Even

collagen coating has been studied in many literatures, these studies focused on permanent

biomedical metal or alloys like titanium [25,26], Ti6Al4V [27], cobalt-based alloys [23]

or stainless steel 316L [28]. To the authors’ knowledge, this is the first report on the effect

of collagen coating on Fe30Mn bioresorbable alloys toward understanding the

interactions of bone marrow stromal cells (BMSCs) with Fe-Mn alloys. We also believe

This article is protected by copyright. All rights reserved.

 this work brings the first clue on how collagen coating effects corrosion behavior in such

resorbable alloys and how it could counterbalance the fact that ion metals leaching out

brings cytotoxicity concerns for the application of such resorbable alloys in temporary
Accepted Article
implants.

2. Materials and Methods

2.1 Sample Fabrication

Figure 1 shows the schematic of alloy fabrication of non-porous and porous Fe30Mn

alloy substrates. A mixture of iron and manganese powders in a weight ratio of 7:3 was

mixed and spun in a high-speed mixer at 800 rpm for 15 mins. The median powder sizes

for iron (IRON325, Chemical Store, New Jersey) and manganese (MN-101, Atlantic

Equipment Engineers, New Jersey) were 44 µm and 10 µm, respectively, according to the

vendor specification. Ethanol with 10wt% of the total mixture was added for lubrication.

The mixture was spun at 800 rpm for 15 mins and cold pressed at 600 MPa using a

hydraulic press (Carver, New Jersey) to obtain coupon compacts with a green body

diameter of 12.7 mm and thickness of 1.3 mm for all tests. The high compaction pressure

we used in this study is to increase the strength of the scaffold and to reduce the

shrinkage after sintering.

To prepare non-porous Fe30Mn samples, half of the mixture was aliquoted and cold

pressed. To prepare porous Fe30Mn samples, the other half of the mixture was mixed

with 10-vol% of ammonium bicarbonate, spun with an addition of 10-wt% of ethanol,

This article is protected by copyright. All rights reserved.

 and then aliquoted and followed by cold pressed. In the sintering process, the green

compacts for porous Fe30Mn samples were loaded in a muffle furnace, heat treated at

120°C for 2 hours, subsequently treated at 1200°C for 3 hours, and finally cooled to room
Accepted Article
temperature. The green compacts for non-porous Fe30Mn samples were sintered at

1200°C for 3 hours and cooled to room temperature. All samples were sintered with a

heating rate of 10°C/min under argon atmosphere. After sintering, all Fe30Mn samples

were mechanically ground with silicon carbide paper in a grit size of 2000, ultrasonically

cleaned in ethanol for 10 minutes, and stored in a desiccator for further testing.

For the reference Ti6Al4V samples, standard Grade 5 Ti6Al4V (ASM Aerospace

Specification Metals; Florida) rods were cut and ground to obtain a disk-shaped sample

size of 12.7 mm in diameter and 1.3 mm in thickness. The Ti6Al4V samples were also

mechanically ground to 2000 grit, ultrasonically cleaned in ethanol, and stored in a

desiccator for further testing.

2.2 Collagen Coating Solution Preparation

The collagen-fibril solution was prepared according to the study of Loo et. al. [29]. In

brief, monomer Type I collagen solution (Advanced Biomatrix; California) was added to

a buffer solution containing deionized water, 200 mM sodium phosphate dibasic adjusted

to pH 7 with hydrochloric acid, and 400 mM potassium chloride, to achieve a

concentration of ~0.3 mg/ml monomer collagen solution. A schematic of collagen coating

preparation is displayed in Figure 3.

This article is protected by copyright. All rights reserved.

 2.2.1 Collagen Coating via Physical Drop Deposition
Accepted Article
The solution containing monomer collagen was placed in a heat block maintaining at

37˚C for at least 6 hours for the assembly of fibrillar collagen. After collagen fibrils

formed in solution, porous Fe30Mn samples were coated by dropping the solution

“freely” on top of the alloy surface and allowed the solution to dry overnight so that only

collagen fibrils remained on the alloy surface. We named this technique physical drop

deposition (DD). Table 1 the optimization parameters for the DD coating of Type I

collagen on porous Fe30Mn substrates. Infrared absorption spectra of the coated samples

prepared from Table 1 were obtained using a Fourier-transform infrared spectroscopy

(FTIR; Spectrum 100). Surface microstructures and compositions of the coated samples

were examined using an optical microscope and a scanning electron microscope (SEM;

FEI Philips XL-40) coupled with energy dispersive x-ray spectroscopy (EDS).

2.2.2 Collagen Deposition via Spin Coating

In another set of experiments, porous Fe30Mn samples were coated with Type I collagen

using the spin coating (SC). These experiments were performed to eliminate the

interference of buffer solution on samples, which was observed for the physical drop

deposition. Table 2 shows the spin coating parameters. Every Fe30Mn porous sample was

centered on a spin coater and 100 µl of collagen fibrils solution was dispensed on the

substrate carefully to achieve a full coverage. After spin coating, samples were examined

This article is protected by copyright. All rights reserved.

 using the SEM. Based on the homogenous collagen coating observed in SC2 and SC3

groups, four additional sample groups were spun coated at a spin speed of 4000, 4250,

4500, and 5000 rpm, respectively, for 30 seconds. These additional samples were
Accepted Article
examined in a cell culture test after one day of incubation. The spin speed of 4250 rpm

for 30 seconds was selected as final spin coating parameters for both experimental

Fe30Mn and control Ti6Al4V samples for further testing.

2.3 Potentiodynamic Polarization Test

Four experimental groups, non-porous (NP-NC), porous (10P-NC), collagen-coated

non-porous (NP-C), and collagen-coated porous (10P-C) Fe30Mn samples were prepared

by the methods described in Sections 2.1 and 2.2. The potentiodynamic corrosion testing

was conducted using the same method demonstrated in the previous study [15]. In brief,

every sample was mounted, cleaned with DI water, and exposed to the electrolyte

McCoy’s 5A (ATCC, Virginia) at 37 ± 2 °C. The current at the sample was recorded when

the sample was applied to the potential scan from -0.15 V to + 0.15 V at a rate of 0.1667

mV/s. Using the Tafel curve, the corrosion potential and corrosion current were

determined. Consequently, the corrosion rate of each sample was calculated using the

following equation:

𝑖𝑖𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 ×𝐸𝐸𝐸𝐸
Corrosion rate (mmpy) = 3.27 × 10−3 𝐷𝐷
--- Eq. 1

, where EW is equivalent weight of the specimen in g, 𝑖𝑖𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 is corrosion current density

in uA/cm2, and D is density of the specimen in g/cm3.

This article is protected by copyright. All rights reserved.

 The average surface area containing porosity of each group was calculated using ImageJ

[30] with adjusted binary threshold. The surface area (SA) of each sample was calculated

in the following equation:

Accepted Article
1
SA (mm2) = 2 𝜋𝜋𝑑𝑑 2 + 𝜋𝜋𝜋𝜋𝜋𝜋 + 𝜋𝜋𝜋𝜋𝜋𝜋 --- Eq. 2

, where 𝑑𝑑 is diameter of the specimen in mm, 𝑡𝑡 is thickness of the specimen in mm, 𝐼𝐼

is average pore depth in mm, and 𝑝𝑝 is sum of pore feret diameter in mm.

2.4 In-Vitro Cell Culture

Four experimental Fe30Mn groups (NP-NC, 10P-NC, NP-C, 10P-C) and two control
Ti6Al4V groups (CT-NC, CT-C) were tested in the direct cytotoxicity test. All samples
were exposed to UV light for 1.5 hours (1 hour for the seeding surface and 0.5 hour for
the opposite side) for sterilization. Each sample was placed in a 24-well plate and rinsed
with cell culture media (McCoy’s 5A supplemented with 10% fetal bovine serum, 5%
Penicillin Streptomycin, and 0.5% Fungizone) prior to cell seeding. Mouse bone marrow
stromal cells (BMSC; ATCC, D1 ORL UVA) within passage number 10-15 were seeded
on each sample in a density of 5×104 cells/well and the cell-scaffold composite was
subsequently placed inside an incubator mimicking the physiological environment (pH =
7, 5% CO2, 37.5 ± 0.5°C) for 1, 3, and 7 days.

This article is protected by copyright. All rights reserved.

 2.4.1 Direct Cytotoxicity Test

After one day of incubation, the cytotoxicity level was evaluated by conducting a
Accepted Article
colorimetric cytotoxicity assay kit (Promega, CytoTox96® Non-Radioactive Cytotoxicity

Assay). The assay measured quantitatively the amount of lactate dehydrogenase (LDH) in

the culture supernatants by measuring the optical density (OD) reading of a formazan

product, which was the reduction form of a tetrazolium compound [iodonitro-tetrazolium

violet; INT] using a VersamaxTM absorbance microplate reader (Molecular Devices;

California). The quantity of the formazan product was proportional to the number of

lysed cells on OD at 490nm. According to the manufacturer’s protocol, the toxicity of

every sample was calculated by the following equation:

Experimental LDH Release (OD)

𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 = 100 × − − − 𝐸𝐸𝐸𝐸. 1
Maximum LDH Release (OD)

, where maximum LDH release is the absorbance of lysis solution provided from the kit.

2.4.2 Fluorescent Staining

After 1, 3, and 7 days of incubation, the cellular morphology and viability were observed

by labelling plasma membrane with 1X of CellMaskTM Plasma Membrane Stain (Life

Technologies, Grand Island, NY) and nuclei with 0.1 ug/ml of Hoechst Fluorescent Stain

(ThermoFisher Scientific, Waltham, MA). Samples were stained for 30 minutes and

rinsed with Live Cell Imaging Solution (ThermoFisher Scientific, Waltham, MA) three

times to wash the excessive dye. The samples were examined using a fluorescent

This article is protected by copyright. All rights reserved.

 microscope where the plasma membrane and nuclei of cells were observed in

fluorescence emission spectra of green and blue, respectively. The cell count and size for

each group were measured using ImageJ where the image sample size was 20 (n = 20)
Accepted Article
and the count was normalized to the number of cells per square millimeter.

2.4.3 Sample Processing for SEM

On the third day of incubation, the experimental groups were rinsed with PBS three times

and filled with 2.5% glutaraldehyde for fixation. The samples were then processed into

DI water rinse, 1% osmium tetroxide, DI water rinse, ethanolic dehydration, and critical

point drying as the final step. Consequently, samples that preserved cellular morphology

were sputter coated with platinum and examined using a SEM microscope (Nova

NanoSEMTM; FEI).

2.5 Statistical Analysis

A minimum of three replicates of the sample group (n = 3) were conducted for every
experiment. Quantitative data from each test were collected and compared using
Student’s t-test. The differences between the studied groups were evaluated through a
one-way analysis of variance (ANOVA) in a statistical analysis software (jmp). A
threshold of α = 0.05 was set to determine statistical significance.

This article is protected by copyright. All rights reserved.

 3. Results

3.1 Sample Characterizations

Accepted Article
The SEM image of non-porous and porous Fe30Mn sample before coating is displayed in

Figure 2. The pore size of the Fe30Mn scaffold was selected between 355-500 µm for the

purpose of bone ingrowth [31].

3.1.1 Drop Deposition Coating

Infrared absorption spectrum of Type 1 collagen on which Fe30Mn substrates coated was

displayed in Figure 4. All samples prepared by different coating parameters listed in

Table 1 showed consistent amide I band and amide II band peaks at 1637 cm-1 and 1546

cm-1 wavenumbers, respectively. Observation of both amide I and amide II band peaks in

this study follows observation of those in references [32,33].

Since similar edge morphology of each group was observed using SEM, only the edging

surface and chemical composition of DD2 (6 hours of fibril collagen self-assembled time

and 90 μl of drop solution drying at 25 ˚C) was displayed as representative in Figure 5.

At the top-left corner of the SEM image in Figure 5, the edge of the sample was iron (Fe)

and manganese (Mn) rich, representing the original Fe30Mn substrate. Toward the edge

of the sample, cracks or fragments of bumps were observed. Those bumps were

particularly sodium (Na) and phosphorus (P) rich, which were the main elements in the

This article is protected by copyright. All rights reserved.

 buffer solution that contained the fibrillar collagen. Since bumps on Fe30Mn samples

were observed after coating, and those bumps resulted from the collagen buffer solution,

we concluded that the excess of buffer solution interacted with the surface of Fe30Mn
Accepted Article
sample during the overnight drying process.

3.1.2 Spin Coating

Figure 6 showed SEM images of collagen-coated Fe30Mn samples using the spin coating

technique. Collagen was observed in all samples prepared by different spinning time and

speeds according to Table 2. However, homogeneous collagen bundles were only

observed in SC2 (spun at a spinning speed of 5000 rpm for 30 seconds) and SC3 (spun at

a spinning speed of 4000 rpm for 30 seconds) samples. At a high spin speed of 6000 rpm,

bundles of fibrillar collagen were observed in Figure 6. (a). In Figure 6. (d), amorphous

“coral” and “mat” structures of collagen were observed for the sample spun at a

combination of low spin speed of 3000 rpm and high spin speed of 6000 rpm.

To identify the ideal spin coating parameters that are most likely to promote cellular
viability, a cell culture test for porous Fe30Mn samples spun coated at 4000, 4250, 4500,
and 5000 rpm for 30 seconds was conducted. In Figure 4, fluorescent images of bone
marrow stems cells (BMSCs) are visible on Fe30Mn samples. Here, the Fe30Mn samples
spun coated at 4250 rpm for 30 seconds showed the highest cellular activity of the group.
For the Fe30Mn sample spun coated at 4250 rpm, BMSCs extended their dendrites
(plasma membrane labeled in green) out to communicate with the other cells, indicating
promising cellular interaction.

This article is protected by copyright. All rights reserved.

 3.2 Potentiodynamic Polarization Test

Once the coating parameter was determined, the electrochemical corrosion test was
Accepted Article
conducted to examine the effect of collagen coating on the corrosion rate of Fe30Mn

samples. Figure 8 showed the representative Tafel curves of experimental Fe30Mn

groups: Fe30Mn without collagen coating (NP-NC), porous Fe30Mn without collagen

coating (10P-NC), Fe30Mn with collagen coating (NP-C), and porous Fe30Mn with

collagen coating (10P-C). The average corrosion potential (Ecorr), corrosion current

density (Icorr), Tafel constants (βa and βc) and linear polarization resistance (Rp) derived

from the Stern-Geary Equation, and corrosion rate (CR) for the experimental Fe30Mn

groups were also listed in Table 3. Average corrosion rate ranked from high to low was

10P-NC > 10P-C > NP-NC > NP-C. Although corrosion rates of collagen-coated samples

were lower than those of non-collagen coated ones, this decrease in corrosion rate was

not significant at the 0.05 level, indicating coating collagen did not have significant effect

on corrosion rate for Fe30Mn samples.

3.3 Direct In-Vitro Test

For the cytotoxicity test, four experimental Fe30Mn groups (NP-NC, 10P-NC, NP-C, and

10P-C) as well as two control Ti6Al4V groups (CT-NC and CT-C) were investigated after

one day of incubation with BMSCs.

This article is protected by copyright. All rights reserved.

 3.3.1 Cytotox96 Cytotoxicity Test

After 1 day of incubation, the cytotoxicity effect of the four experimental Fe30Mn groups
Accepted Article
and two control Ti6Al4V groups on BMSCs were evaluated and compared. The

numerical average and standard deviation values of percent cytotoxicity for each group

were listed in Table 4 and plotted in Figure 9. The average cytotoxicity level was 14.97 %

for the non-coated porous group (10P-NC), representing the highest level of cytotoxicity

effect, and 1.80 % for the coated non-porous group (NP-C), representing the lowest level

of cytotoxicity effect. A significant decrease in the cytotoxicity level from the non-coated

porous group (10P-NC) to the coated porous group (10P-C) was found. There was no

significant difference between the non-porous Fe30Mn groups (NP-NC and NP-C) in the

cytotoxicity level, neither was the significant difference between the control Ti6Al4V

groups (CT-NC and CT-C).

3.3.2 Cellular Behavior

Qualitative evaluation of cell morphology, detachment, and membrane integrity for each

experimental group were examined after 1, 3 and 7 days of incubation, as displayed in

Figure 10. On day 1, cells appeared to be the healthiest on the coated non-porous group

(NP-C) and the unhealthiest on the non-coated porous group (10P-NC). On day 3,

BMSCs seeded on the collagen-coated Fe30Mn groups (NP-C and 10P-C) were grown to

achieve a higher confluency than those seeded on the non-coated groups (NP-NC and

10P-NC), respectively. As the culture time reached to day 7, BMSCs were grown to

100% confluency for all Fe30Mn groups except for the 10P-NC group.

This article is protected by copyright. All rights reserved.

 Figure 11 shows the average cell count (a) and size (b) after 1 and 3 days of incubation.

The corresponding numerical values are listed in Table 4. Compared to the non-coated

Fe30Mn groups, the collagen-coated Fe30Mn groups showed significant increases in the
Accepted Article
average cell count and size within 24 hours. After 3 days of incubation, there was no

significant difference in the cell count between the coated non-porous Fe30Mn group

(NP-C) and the control Ti6Al4V groups (CT-NC and CT-C).

Figure 12 shows the qualitative SEM images of cells on each experimental group after 3

days of incubation. Cells adhered to the NP-NC, NP-C and 10P-C groups appeared

flattened, spreading thin extensions to reach out other cells. For the 10P-NC group, cells

sat erect on the surface of scaffold, indicating loose-attachment. Further comparison of

cell morphology on the 10P-NC and 10P-C groups is displayed in Figure 13. No cell was

observed at the edge of the pore on the 10P-NC sample (Figure 13. (a)), while flattened

cells were observed surrounding and even covering the entire pore (Figure 13. (c)). The

profound difference in cell morphology between the porous non-collagen-coated group

(10P-NC) and porous collagen-coated group (10P-C) demonstrated that the collagen

coating was necessary for porous Fe30Mn to make cells show healthy discrete

intracytoplasmatic granules.

4. Discussion

In the previous study [34] we invested Fe30Mn with 5%, 10%, and 60% porosity on alloy

mechanical properties, degradation rate and biocompatibility. In short, both the average

ultimate compressive strengths for non-porous and porous Fe30Mn were above 700 MPa,

This article is protected by copyright. All rights reserved.

 which were higher than a typical human wet compact bone. The Fe30Mn scaffolds with

10-vol% porosity are considered promising resorbable scaffolds based on the results of

compression tests, corrosion experiments and cytotoxicity studies. Our previous work
Accepted Article
also showed that there is promising biocompatibility of Fe30Mn that was implanted

in-vivo into rats and bone growth at the interface was noted [35].

Fe30Mn samples coated by fibrillar Type I collagen using the drop deposition technique

showed similar absorption IR spectrum containing both amide I and amide II band peaks.

Detection of amide I and amide II peaks on all collagen-coated samples using the drop

deposition technique demonstrated that collagen adsorbed on the alloy substrate

successfully. Nevertheless, flaky cracks appeared close to the edge of the sample after

coating and drying overnight. The authors initially suspected the cracks were metal

hydroxides [36,37] resulting from the electrochemical reactions between the alloy and

buffer solution. However, EDS results showed that the cracks were sodium and phosphate

rich instead of iron and manganese rich, suggesting that the occurrence of cracks was

merely caused by the buffer solution. The excess of fibrillar collagen buffer solution

returned salts to their anhydrous forms after allowing the solution to dry naturally,

leaving segregated salts on the surface of the sample.

Since samples prepared from drop deposition showed inhomogeneous coating due to

excess of buffer solution, spin coating was used to quickly remove the buffer solution and

obtain homogenous collagen Type I coverage. Spin speed at 3000 rpm, in most cases, is

the starting reference speed for trial and error [38]. Niwa, et. al. spin coated collagen on

their polymer PLLA nanosheet at 4000 rpm for 20 seconds and showed supported cell

adhesive properties [39]. In the paper of Saeidi, et. al., no fibrillar alignment should be

This article is protected by copyright. All rights reserved.

 expected at spin speed lower than 3000 rpm [40]. Based on the above research papers, we

reasoned our experimental spinning speed starting at 3000 rpm and above within 30

seconds.
Accepted Article
Although spin coating prevented salt residues remaining on samples, coating uniformity

was complex and difficult to achieve. For samples spun at high spin speed, the centripetal

force pulled collagen solution flow toward the edge of the sample, which made few

collagen molecules stay in the middle of the sample. Reducing the spin speed can keep

collagen molecules in the middle, yet the coating may not cover the entire substrate

because of poor centripetal force and the crystallization of solution that may occur [41].

The authors chose spin speed at 4250 rpm as the final coating speed, which offered the

most uniform coating and high cellular response.

For the electrochemical corrosion test, the corrosion potentials after coating shifted more

positive for both non-porous and porous Fe30Mn, as shown in Table 3, because the

collagen coating acted as a protective layer that slowed down the anodic reactions for the

alloy, which consequently reduced the metal dissolution releasing rate of electrons. We

published data for the immersion test for non-porous and porous Fe30Mn without coating

in our previous paper [15]. Since we are only looking at trends here, and the trends in the

immersion test match those of the potentiodynamic polarization test, we presented the

polarization test only here. The electrochemical corrosion test also showed a reduction in

corrosion rates for both non-porous and porous collagen-coated samples comparing to

samples without collagen coating (NP-C < NP-NC; 10P-NC < 10P-C) in Table 3.

Collagen coating reduced the corrosion rate of Fe30Mn because it prevented the direct

exposure of alloy surface to the electrolyte, hence reducing oxidation. Nevertheless, there

This article is protected by copyright. All rights reserved.

 was no statistically significant difference in corrosion rates between the collagen-coated

and non-coated groups, demonstrating that collagen coating only minimally hindered the

overall corrosion rates of Fe30Mn.

Accepted Article

Collagen coating of the Fe-Mn alloys significantly reduced the cytotoxicity level of the

porous Fe30Mn groups (10P-C < 10P-NC), as shown in Table 4 and Figure 9. The

coating served as an intermediate layer between cells and the substrate. It contained high

level of amino acids including glycine, proline, and hydroxyproline [42] that provided

binding sites to cell-surface receptors or integrins on the cell membrane, which greatly

reduced the occurrence of anoikis or cell-detachment-apoptosis. Consequently, increasing

number of binding sites facilitated cell attachment, metabolic activity and mobility which

enabled cell-to-cell communication. This was demonstrated by the significant increase of

the day 1 average cell size, which was proportional to the cell adhesion, from the

non-coated groups (NP-NC, 10P-NC, and CT-NC) to the collagen-coated groups (NP-C,

10P-C, and CT-C), respectively (Figure 11b and Table 4). Furthermore, the average cell

count values for the collagen coated Fe30Mn groups were significantly higher than those

for the non-coated Fe30Mn groups, indicating that the collagen coating improved the

initial cellular viability on Fe30Mn. The average cell size decreased from day 1 to day 3

for the collagen-coated Fe30Mn groups because the cells nearly reached full confluency,

squeezing themselves to cover the entire surface. On the other hand, the average cell size

for the non-coated Fe30Mn groups at day 3 increased because the cells still had room to

extend their dendrites to communicate to the other cells.

This article is protected by copyright. All rights reserved.

 The porous Fe30Mn groups (10P-NC and 10P-C) showed a higher toxicity level than the

non-porous Fe30Mn groups (NP-NC and NP-C) because a higher concentration level of

metal ions (i.e. Fe2+/3+ and Mn2+) leached into the environment which eventually lead to
Accepted Article
cell necrosis. Even the average cytotoxicity level for the 10P-NC group showed the

highest among those for all groups, however, it was still within 30% of the average

cytotoxicity level for the negative control Ti6Al4V group, and according to the standard

ISO10993-5 [43], the 10P-NC group was not considered severe cytotoxic. Although the

coated non-porous group (NP-C) showed higher cell viability and adhesion than the

coated porous group (10P-C), the coated porous group was more desirable because it

provided three-dimensional environment for cells to facilitate their phenotypes and

cell-to-cell interactions, mimicking the physiological environment closer to tissues and

organs [44,45]. Since there was no statistically significant difference between the 10P-C

and 10P-NC group in the corrosion test, yet the 10P-C group showed a significant lower

cytotoxicity level than the 10P-NC group in the cytotoxicity test, we can conclude that

collagen coating is a valid route towards increasing the osteointegration for porous

Fe30Mn without hindering the alloy degradation rate.

5. Conclusions

Fe30Mn is a promising biodegradable material for transient orthopedic fixation devices,

but further improvements in osteointegration of these materials are necessary. The spin

coating technique was used to obtain homogeneous coating of fibrillar Type I collagen on

the surface of Fe-Mn alloys with different porosity levels. According to the

This article is protected by copyright. All rights reserved.

 potentiodynamic polarization test, no significant differences between the corrosion rates

of the collagen-coated groups (NP-C and 10P-C) and the non-coated groups (NP-NC and

10P-NC) were identified in this study. At the same time, the cytotoxicity test showed that
Accepted Article
collagen coating of the Fe-Mn alloys significantly reduced the cytotoxicity level of the

porous samples after 1 day of incubation. The quantitative results of the average

corrosion rates and cytotoxicity levels, combined with the qualitative results of cell count

values, sizes, and morphology obtained from the fluorescence microscopy and SEM

images, showed that the collagen coating improved the osteointegration of the Fe30Mn

alloys.

References

[1] M. Niinomi, M. Nakai, J. Hieda, Development of new metallic alloys for

biomedical applications, Acta Biomater. 8 (2012) 3888–3903.

doi:10.1016/j.actbio.2012.06.037.

[2] X. Li, X. Liu, S. Wu, K.W.K. Yeung, Y. Zheng, P.K. Chu, Design of magnesium

alloys with controllable degradation for biomedical implants: From bulk to

surface, Acta Biomater. 45 (2016) 2–30. doi:10.1016/J.ACTBIO.2016.09.005.

[3] H. Hermawan, D. Dubé, D. Mantovani, Developments in metallic biodegradable

stents, Acta Biomater. 6 (2010) 1693–1697. doi:10.1016/J.ACTBIO.2009.10.006.

[4] C. Gao, S. Peng, P. Feng, C. Shuai, Bone biomaterials and interactions with stem

cells, Nat. Publ. Gr. 5 (2017). doi:10.1038/boneres.2017.59.

This article is protected by copyright. All rights reserved.

 [5] R. Gorejová, L. Haverová, R. Oriňaková, A. Oriňak, M. Oriňak, Recent

advancements in Fe-based biodegradable materials for bone repair, J. Mater. Sci.

54 (2019) 1913–1947. doi:10.1007/s10853-018-3011-z.

Accepted Article
[6] Y. Jones-Hall, G. Breur, L. Stanciu, M. Heiden, E. Nauman, M. Traverson, In Vivo

Evaluation of Biodegradability and Biocompatibility of Fe30Mn Alloy, Vet. Comp.

Orthop. Traumatol. 31 (2018) 010–016. doi:10.3415/vcot-17-06-0080.

[7] M.P. Staiger, A.M. Pietak, J. Huadmai, G. Dias, Magnesium and its alloys as

orthopedic biomaterials : A review, Biomaterials. 27 (2006) 1728–1734.

doi:10.1016/j.biomaterials.2005.10.003.

[8] F. Witte, V. Kaese, H. Haferkamp, E. Switzer, A. Meyer-Lindenberg, C.J. Wirth, H.

Windhagen, In vivo corrosion of four magnesium alloys and the associated bone

response, Biomaterials. 26 (2005) 3557–3563.

doi:10.1016/j.biomaterials.2004.09.049.

[9] J. He, F. He, D. Li, Y. Liu, Y. Liu, Y. Ye, D. Yin, Advances in Fe-based

biodegradable metallic materials, RSC Adv. 6 (2016) 112819–112838.

doi:10.1039/c6ra20594a.

[10] E. Mouzou, C. Paternoster, R. Tolouei, A. Purnama, P. Chevallier, D. Dubé, F.

Prima, D. Mantovani, In vitro degradation behavior of Fe – 20Mn – 1 . 2C alloy in

three different pseudo-physiological solutions, Mater. Sci. Eng. C. 61 (2016) 564–

573. doi:10.1016/j.msec.2015.12.092.

This article is protected by copyright. All rights reserved.

 [11] M. Schinhammer, A.C. Hänzi, J.F. Löffler, P.J. Uggowitzer, Design strategy for

biodegradable Fe-based alloys for medical applications, Acta Biomater. 6 (2010)

1705–1713. doi:10.1016/j.actbio.2009.07.039.
Accepted Article
[12] J. Hufenbach, H. Wendrock, F. Kochta, U. Kühn, A. Gebert, Novel biodegradable

Fe-Mn-C-S alloy with superior mechanical and corrosion properties, Mater. Lett.

186 (2017) 330–333. doi:10.1016/j.matlet.2016.10.037.

[13] N. Mohd, N. Boon, A. Hakim, F. Adibah, A. Majid, H. Hermawan, Degradation

and in vitro cell-material interaction studies on hydroxyapatite-coated

biodegradable porous iron for hard tissue scaffolds, J. Orthop. Transl. 2 (2014)

177–184. doi:10.1016/j.jot.2014.07.001.

[14] M. Heiden, E. Nauman, L. Stanciu, Bioresorbable Fe – Mn and Fe – Mn – HA

Materials for Orthopedic Implantation : Enhancing Degradation through Porosity

Control, Adv. Healthc. Mater. 6 (2017) 1700120. doi:10.1002/adhm.201700120.

[15] S.M. Huang, E.A. Nauman, L.A. Stanciu, Investigation of porosity on mechanical

properties, degradation and in-vitro cytotoxicity limit of Fe30Mn using space

holder technique, Mater. Sci. Eng. C. 99 (2019) 1048–1057.

doi:10.1016/J.MSEC.2019.02.055.

[16] L.N. Luong, J. Ramaswamy, D.H. Kohn, Effects of osteogenic growth factors on

bone marrow stromal cell differentiation in a mineral-based delivery system,

Biomaterials. 33 (2012) 283–294. doi:10.1016/j.biomaterials.2011.09.052.

This article is protected by copyright. All rights reserved.

 [17] H.S. Alghamdi, R. Bosco, J.J.J.P. Van Den Beucken, X.F. Walboomers, J.A.

Jansen, Osteogenicity of titanium implants coated with calcium phosphate or

collagen type-I in osteoporotic rats, Biomaterials. 34 (2013) 3747–3757.

Accepted Article
doi:10.1016/j.biomaterials.2013.02.033.

[18] Y. Förster, R. Bernhardt, V. Hintze, S. Möller, M. Schnabelrauch, D. Scharnweber,

S. Rammelt, Collagen / glycosaminoglycan coatings enhance new bone formation

in a critical size bone defect — A pilot study in rats, Mater. Sci. Eng. C. 71 (2017)

84–92. doi:10.1016/j.msec.2016.09.071.

[19] Y. Zhukova, C. Hiepen, P. Knaus, M. Osterland, S. Prohaska, J.W.C. Dunlop, P.

Fratzl, E. V. Skorb, The Role of Titanium Surface Nanostructuring on

Preosteoblast Morphology, Adhesion, and Migration, Adv. Healthc. Mater. 6

(2017) 1601244. doi:10.1002/adhm.201601244.

[20] Y. Sun, C.S. Chen, J. Fu, Forcing stem cells to behave: a biophysical perspective of

the cellular microenvironment., Annu. Rev. Biophys. 41 (2012) 519–42.

doi:10.1146/annurev-biophys-042910-155306.

[21] H.T. Aiyelabegan, E. Sadroddiny, Fundamentals of protein and cell interactions in

biomaterials, Biomed. Pharmacother. 88 (2017) 956–970.

doi:10.1016/J.BIOPHA.2017.01.136.

[22] M.D. Shoulders, R.T. Raines, Collagen Structure and Stability, Annu Rev

Biochem. 78 (2010) 929–958.

doi:10.1146/annurev.biochem.77.032207.120833.COLLAGEN.

This article is protected by copyright. All rights reserved.

 [23] R. Müller, J. Abke, E. Schnell, D. Scharnweber, R. Kujat, C. Englert, D. Taheri, M.

Nerlich, P. Angele, Influence of surface pretreatment of titanium- and cobalt-based

biomaterials on covalent immobilization of fibrillar collagen, Biomaterials. 27

Accepted Article
(2006) 4059–4068. doi:10.1016/j.biomaterials.2006.03.019.

[24] S. Rammelt, E. Schulze, R. Bernhardt, U. Hanisch, D. Scharnweber, H. Worch, H.

Zwipp, A. Biewener, Coating of titanium implants with type-I collagen, J. Orthop.

Res. 22 (2004) 1025–1034.

[25] D.G. Costa, E.P. Ferraz, R.P.F. Abuna, P.T. De Oliveira, M. Morra, M.M. Beloti,

A.L. Rosa, The effect of collagen coating on titanium with nanotopography on in

vitro osteogenesis, J. Biomed. Mater. Res. - Part A. (2017) 2783–2788.

doi:10.1002/jbm.a.36140.

[26] X. Yu, M. Wei, J. Walsh, Covalent Immobilization of Collagen on Titanium

through Polydopamine Coating to Improve Cellular Performances of MC3T3-E1

Cells, RSC Adv. 4 (2014) 7185–7192. doi:10.1039/c3ra44137g.

[27] Y. Yang, S. Yang, Y. Wang, Z. Yu, H. Ao, H. Zhang, L. Qin, O. Guillaume, D.

Eglin, R. Geoff Richards, T. Tang, Anti-infective efficacy, cytocompatibility and

biocompatibility of a 3D-printed osteoconductive composite scaffold

functionalized with quaternized chitosan., Acta Biomater. 46 (2016) 112–128.

doi:10.1016/j.actbio.2016.09.035.

[28] Q. Lin, X. Ding, F. Qiu, X. Song, G. Fu, J. Ji, In situ endothelialization of

intravascular stents coated with an anti-CD34 antibody functionalized

This article is protected by copyright. All rights reserved.

 heparin-collagen multilayer, Biomaterials. 31 (2010) 4017–4025.

doi:10.1016/j.biomaterials.2010.01.092.
Accepted Article
[29] R.W. Loo, J.B. Goh, C.C.H. Cheng, N. Su, M.C. Goh, In vitro Synthesis of Native

, Fibrous Long Spacing and Segmental Long Spacing Collagen, (2012).

doi:10.3791/4417.

[30] C.A. Schneider, W.S. Rasband, K.W. Eliceiri, NIH Image to ImageJ: 25 years of

image analysis, Nat. Methods. 9 (2012) 671. https://doi.org/10.1038/nmeth.2089.

[31] a. Laptev, O. Vyal, M. Bram, H.P. Buchkremer, D. Stöver, Green strength of

powder compacts provided for production of highly porous titanium parts, Powder

Metall. 48 (2005) 358–364. doi:10.1179/174329005X73838.

[32] B. De Campos Vidal, M.L.S. Mello, Collagen type I amide I band infrared

spectroscopy, Micron. 42 (2011) 283–289. doi:10.1016/j.micron.2010.09.010.

[33] K. Belbachir, R. Noreen, G. Gouspillou, C. Petibois, Collagen types analysis and

differentiation by FTIR spectroscopy, Anal. Bioanal. Chem. 395 (2009) 829–837.

doi:10.1007/s00216-009-3019-y.

[34] S. Pal, Mechanical Properties of Biological Materials, in: Des. Artif. Hum. Joints

Organs, Springer US, Boston, MA, 2014: pp. 23–40.

doi:10.1007/978-1-4614-6255-2_2.

This article is protected by copyright. All rights reserved.

 [35] M. Traverson, M. Heiden, L. Stanciu, E. Nauman, Y. Jones-Hall, G. Breur, In Vivo

evaluation of biodegradability and biocompatibility of Fe30Mn alloy, Vet. Comp.

Orthop. Traumatol. 31 (2018) 010–016. doi:10.3415/VCOT-17-06-0080.

Accepted Article
[36] E. Zhang, H. Chen, F. Shen, Biocorrosion properties and blood and cell

compatibility of pure iron as a biodegradable biomaterial, J. Mater. Sci. Mater.

Med. 21 (2010) 2151–2163. doi:10.1007/s10856-010-4070-0.

[37] Y.F. Zheng, X.N. Gu, F. Witte, Biodegradable metals, Mater. Sci. Eng. R Reports.

77 (2014) 1–34. doi:10.1016/j.mser.2014.01.001.

[38] P.C. Sukanek, Dependence of Film Thickness on Speed in Spin Coating, J.

Electrochem. Soc. 138 (1991) 1712–1719. doi:10.1149/1.2085860.

[39] D. Niwa, T. Fujie, T. Lang, N. Goda, S. Takeoka, Heterofunctional nanosheet

controlling cell adhesion properties by collagen coating, J. Biomater. Appl. 27

(2012) 131–141. doi:10.1177/0885328210394470.

[40] N. Saeidi, E.A. Sander, R. Zareian, J.W. Ruberti, Production of highly aligned

collagen lamellae by combining shear force and thin film confinement, Acta

Biomater. 7 (2011) 2437–2447. doi:10.1016/j.actbio.2011.02.038.

[41] M. Habibi, A. Rahimzadeh, M. Eslamian, On dewetting of thin films due to

crystallization (crystallization dewetting), Eur. Phys. J. E. 39 (2016).

doi:10.1140/epje/i2016-16030-9.

This article is protected by copyright. All rights reserved.

 [42] J.D. Harvey Lodish, Arnold Berk, S Lawrence Zipursky, Paul Matsudaira, David
Baltimore, Collagen: The Fibrous Proteins of the Matrix, in: Mol. Cell Biol., 4th
ed., New York: W. H. Freeman, 2000.
Accepted Article
[43] International Standard ISO 10993-5 Biological evaluation of medical devices -
Part 5: Tests for cytotoxicity: in vitro methods, Int. Stand. ISO. 3 Ed (2009) 42.

[44] A.H. Yusop, A.A. Bakir, N.A. Shaharom, M.R. Abdul Kadir, H. Hermawan, Porous
biodegradable metals for hard tissue scaffolds: A review, Int. J. Biomater. 2012
(2012). doi:10.1155/2012/641430.

[45] C. Gao, Y. Deng, P. Feng, Z. Mao, P. Li, B. Yang, J. Deng, Y. Cao, C. Shuai, S.
Peng, C. Gao, Y. Deng, P. Feng, Z. Mao, P. Li, B. Yang, J. Deng, Y. Cao, C. Shuai,
S. Peng, Current Progress in Bioactive Ceramic Scaffolds for Bone Repair and
Regeneration, Int. J. Mol. Sci. 15 (2014) 4714–4732. doi:10.3390/ijms15034714.

Figures

Figure 1. Schematic illustration of porous Fe30Mn fabrication using the space holder
technique

This article is protected by copyright. All rights reserved.

 Figure 2. Optical and SEM images of Fe30Mn without macro-pore (NP) and with 10-vol%
porosity (10P). Data modified from [15].
Accepted Article

Figure 3. Schematic illustration of collagen coating preparation.

This article is protected by copyright. All rights reserved.

 Figure 4. FTIR spectrum of collagen-coated samples prepared from Table 1: (a) DD1; 6
Accepted Article
hours of fibril collagen self-assembled time and 90 μl of drop solution drying at 4 ˚C, (b) DD2;
6 hours of fibril collagen self-assembled time and 90 μl of drop solution drying at 25 ˚C, (c)
DD3; 12 hours of fibril collagen selfassembled time and 90 μl of drop solution drying at 25 ˚C,
and (d) DD4; 12 hours of fibril collagen selfassembled time and 200 μl of drop solution drying
at 25 ˚C.

This article is protected by copyright. All rights reserved.

 Figure 5. EDS map analysis of DD2 sample (Fe30Mn substrate coated with Type I
collagen; 6 hours of fibril collagen self-assembled time and 90 μl of drop solution drying at 25
˚C). Cracking bumps on the surface of the sample were sodium and phosphorus rich. The edge
of the sample (top-left corner) was iron and manganese rich. The length of each scale bar is 1
Accepted Article
mm.

This article is protected by copyright. All rights reserved.

 Figure 6. SEM images of collagen-coated Fe30Mn samples prepared from Table 2: (a)
SC1; spun at a spinning speed of 6000 rpm for 30 seconds, (b) SC2; spun at a spinning speed of
5000 rpm for 30 seconds, (c) SC3; spun at a spinning speed of 4000 rpm for 30 seconds, and (d)
SC4; spun at a spinning speed of 3000 rpm for 20 seconds and subsequently spun at a spinning
Accepted Article
speed of 6000 rpm for 40 seconds. The length of each scale bar is 3 μm.

This article is protected by copyright. All rights reserved.

 Figure 7. Fluorescent images of bone marrow stem cells (BMSCs) on porous Fe30Mn
samples coated with Type I collagen at the spinning speed of (a) 4000 rpm, (b) 4250 rpm, (c)
4500 rpm, and (d) 5000 rpm after 1 day of incubation. The length of each scale bar is 100 μm.
Accepted Article

This article is protected by copyright. All rights reserved.

 Figure 8. Representative potentiodynamic polarization curves of the experimental Fe30Mn
groups in McCoy’s 5A medium (pH 7.4) at 37 ± 2 °C. The squares represented Fe30Mn without
collagen coating (NP-NC), circles represented Fe30Mn with collagen coating (NP-C), upward
triangles represented porous Fe30Mn without collagen coating (10P-NC), and downward
Accepted Article
triangles represented porous Fe30Mn with collagen coating (10P-C).

This article is protected by copyright. All rights reserved.

 Figure 9. Cytotoxicity level of the studied groups on bone marrow stem cells (BMSCs)
after 1 day of incubation. Studied experimental groups were Fe30Mn without collagen coating
(NP-NC), Fe30Mn with collagen coating (NP-C), porous Fe30Mn without collagen coating
(10P-NC), and porous Fe30Mn with collagen coating (10P-C). Studied control groups were
Accepted Article
Ti6Al4V without collagen coating (CT-NC) and Ti6Al4V with collagen coating (CT-C). The
bracket with the asterisk mark indicated the significant difference between the 10P-NC and
10P-C groups (α < 0.05).

This article is protected by copyright. All rights reserved.

 Figure 10. Fluorescent images of BMSCs on the studied experimental groups: (a-c)
NP-NC, (d-f) NP-C, (g-i) 10P-NC, and (j-l) 10P-C scaffolds after 1 day (a, d g, j), 3 days (b, e,
h, k), and 7 days (c, f, i, l) of incubation. The length of each scale bar is 100 μm.
Accepted Article

This article is protected by copyright. All rights reserved.

 Figure 11. Average bone marrow stem cell count (a) and size (b) distributed on the studied
groups after 1 and 3 days. The bracket with the asterisk mark indicated the significant
differences among the collagencoated and non-coated groups (α < 0.05).
Accepted Article

This article is protected by copyright. All rights reserved.

 Figure 12. SEM images of BMSCs on the experimental groups: (a) NP-NC, (b) NP-C, (c)
Accepted Article
10P-NC, and (d) 10P-C scaffolds after 3 days of incubation. The length of each scale bar is 10
μm.

This article is protected by copyright. All rights reserved.

 Figure 13. SEM images of BMSCs on (a-b) 10P-NC and (c-d) 10P-C scaffolds after 3 days
of incubation at different magnifications. The dashed white boxes inside (a) and (b) were
zoomed to (c) and (d), respectively, to show different morphologies of BMSCs.
Accepted Article

This article is protected by copyright. All rights reserved.

 Tables

Table 1. Collagen coating design for Fe-30Mn substrates using the physical drop
deposition (DD) technique. All coated Fe-30Mn samples contained 10% pore.
Accepted Article
Self-Assembly Time Drying Temperature after Volume of Solution
Sample Group
(hours) Coating (˚C) Drop (μl)

DD1 6 4 90

DD2 6 25 90

DD3 12 25 90

DD4 12 25 200

Table 2. Collagen coating design for Fe-30Mn substrates using the spin coating (SC)
technique. All coated Fe30Mn samples contained 10% pore and were spun with 100 μl of
solution that contained fibril collagen prepared after 12 hours of self-assembled time.

Sample Group Solution Spinning Time (s) Solution Spinning Speed (rpm)

SC1 30 6000

SC2 30 5000

SC3 30 4000

SC4 20/40 3000/6000

This article is protected by copyright. All rights reserved.

 Table 3. Average sample corrosion potentials (Ecorr), corrosion current densities (Icorr),
Tafel constants (βa and βc), linear polarization resistance (Rp), corrosion rates (CR)
Accepted Article
Sample Description Ecorr Icorr βa/βc Rp CR
Group
(V) (µA/cm )2
(mV) (kΩ/cm ) 2
(µmpy)

NP-NC Non-porous, -0.74 ± .06 1.42 ± 1.0 βa: 203 ± 141 42.1 ± 20 19.6 ± 14
no collagen βc: 286 ± 84

10P-NC 10% pore, -0.80 ± .06 2.11 ± 0.5 βa: 226 ± 71 23.6 ± 11 29.8 ± 8
no collagen βc: 217 ± 59

NP-C Non-porous, -0.95 ± .48 1.32 ± 0.6 βa: 275 ± 63 44.4 ± 21 18.2 ± 8
collagen βc: 250 ± 119

10P-C 10% pore, -0.73 ± .02 1.29 ± 0.7 βa: 234 ± 106 49.4 ± 33 18.3 ± 9
collagen βc: 268 ± 24
obtained from the potentiodynamic polarization curves.

Table 4. Average sample cytotoxicity levels after 1 day of incubation as well as average
cell counts and sizes after 1 and 3 days of incubation.

Count (# / mm2) Size (µm)

Cytotoxicity Level
Sample
Group
(%)
Day 1 Day 3 Day 1 Day 3

NP-NC 6.20 ± 2.4 855 ± 277 3622 ± 736 67 ± 16 76 ± 16

NP-C 1.80 ± 1.8 2599 ± 266 6273 ± 454 82 ± 13 56 ± 9

10P-NC 14.97 ± 5.3 650 ± 567 1599 ± 663 54 ± 10 72 ± 22

This article is protected by copyright. All rights reserved.

 10P-C 9.08 ± 3.5 2376 ± 401 3984 ± 673 66 ± 10 38 ± 7

CT-NC 3.56 ± 0.9 703 ± 299 6856 ± 514 57 ± 11 39 ± 8

Accepted Article
CT-C 2.57 ± 2.0 892 ± 180 6712 ± 1098 84 ± 10 40 ± 2.0

This article is protected by copyright. All rights reserved.